To solve the problem of high memory and computational resource demands in obstacle detection models within autonomous driving perception domain controllers, a lightweight obstacle detection method based on improved YOLOv8 was proposed. This method reconstructs the YOLOv8 backbone network using FasterNet, which utilizes less memory access and computational resources. To mitigate the accuracy decline and the insufficient detection capabilities for small objects caused by model lightweighting, three main improvements were made to YOLOv8: SPD-Conv (space-to-depth convolution) was used to replace traditional stride convolution in the neck network to enhance small object feature extraction. IPIoU(inner powerful IoU), combining the concepts of IIoU(inner IoU) and PIoU(powerful IoU), is introduced as the bounding box regression loss to accelerate loss convergence and improve small object detection performance. SimAM (simple attention module) was incorporated to further enhance model detection accuracy. Experimental results demonstrate that, compared to the original model, the improved model achieves a reduction of 29.1% in parameters, 20.5% in computational load, and 28.8% in model size, while increasing mAP@0.5 by 1.2%. Once deployed in autonomous driving vehicle controllers, the model effectively detects obstacles on the road ahead.

It is essential to perform equipment reliability classification in order to devote limited resources to NPP equipment management reasonably, improve equipment reliability and availability while reducing maintenance workload and cost, and enable NPP’s safe, reliable, and economical operation. In light of engineering characteristics of the demonstration fast reactor and the challenges during the construction, an attainable new method was developed to standardize the process of equipment reliability classification and complete the classification of all systems. The differences and characteristics among the methods were compared, the implementation process of the new method was proposed and demonstrated with two real systems. The application shows that the proposed method is efficient, effective and rational, hence can offer to assist other NPPs of similar reactors to implement reliability classification during NPP construction.

A simplified lower-limb exoskeleton model was established for the prototype, and the D-H parameter method was used to perform dynamic analysis. Joint angles were measured experimentally and used as inputs for the controller. To address the robot's trajectory tracking problem, traditional PID control was employed, showing good tracking performance but slow response and parameter tuning speed. Although PSO(particle swarm optimization) accelerated the parameter tuning, issues with low convergence accuracy and local optimum traps persisted. Therefore, a PID control based on a chaotic-mapping improved PSO algorithm was designed. The results show that the randomness was enhanced, the parameter tuning speed was increased, and the tracking error was reduced. Simscape was used for visual simulation of joint angles, and the control performance was further validated through various experiments.

Addressing challenges such as large memory footprint, high computational complexity, and insufficient real-time detection speed in road crack detection models for complex scenarios, a highly efficient and precise algorithm named FCG-YOLO was proposed. Lightweight modules and attention mechanisms were integrated, and traditional feature fusion pyramids were enhanced.The algorithm incorporates PConv into the residual calculation module of YOLOv8n to introduce the improved C2f_Faster structure, thereby reducing model parameters and computational complexity. To enhance detection accuracy, GAM(global attention mechanism) was introduced into the backbone, and the Feature Fusion Pyramid SPPF was improved to SPPFCSPC module, enhancing the model’s ability to represent and fuse features of road cracks.The impact of each module on algorithm performance was verified through ablation experiments, identifying a lightweight and accurate model configuration. Furthermore, the robustness and generalization of the algorithm were explored in practical application scenarios.FCG-YOLO demonstrates outstanding detection efficiency, achieving a detection accuracy of 90.3% mAP50 and 74.4% mAP50-95 on the validation set, with a detection speed of 345 frames per second. These results highlight its high detection efficiency and significant practical value.

In order to study the prediction and health management of PEMFCs(proton exchange membrane fuel cells) for vehicles, a method combining GWO(grey wolf optimizer) and RBF(radial basis function) neural network with relative power loss rate as a health indicator was proposed to predict the remaining useful life of vehicular PEMFCs. Firstly, by analyzing the polarization curve of the fuel cell at the initial moment, a calculation method based on the relative power loss rate as a health indicator was constructed, and its feasibility was verified using the grey correlation analysis method. Then, the RBF neural network optimized by GWO algorithm was applied to predict the remaining useful life of vehicular PEMFCs. Finally, the proposed method was validated using two datasets. The results show that compared with other methods, the GWO-RBF method proposed in this paper has the smallest average absolute percentage error and root mean square error, the largest coefficient of determination, and a relative error of less than 1%. It is concluded that the proposed method can be used to predict the remaining useful life of vehicular PEMFCs with fewer datasets and better accuracy.

Frost damage is a key problem in the construction of alpine tunnels. A numerical model was established to consider heat conduction and convection for designing effective thermal insulation measures. Orthogonal test conditions were designed to analyze the significance of different thermophysical parameters on lining temperature and their influence on lining damage. The results show that the convection coefficient, the thermal conductivity of the insulation layer, and the thermal conductivity of the surrounding rock have significant influences. The thermal conductivity of the insulation layer has the most significant impact. As its thermal conductivity increases, the lowest lining temperature decreases, and the damage level increases. An increase in the convection coefficient decreases the minimum lining temperature and increases damage. Higher thermal conductivity of surrounding rock raises the minimum lining temperature, spreading damage from the vault to the sides. Reducing the thermal conductivity and convection coefficient of the insulation layer, while increasing the thermal conductivity of the surrounding rock, improves overall damage distribution.

In order to investigate the heat and mass transfer of liquid droplet impinging on a heated surface with different wettability, VOF(volume of fluid) numerical simulation method was used to analyzed the mechanisms of wall wettability and surface temperature on droplet morphology and heat transfer characteristics. The results show that the hydrophilic wall is favorable for droplet spreading, while the hydrophobic wall is favorable for droplet rebound. With the increase of contact angle, the maximum spreading factor of droplets decreases, the time to reach the maximum spreading factor is shortened, and the average heat flux of the wall surface decreases. Surface temperature has less influence on the droplet spreading stage, with the increase of surface temperature, the droplet phase transition rate accelerates, the average heat flux of the wall surface increases, and the Leidenfrost phenomenon occurs when the surface temperature exceeds the critical temperature.

In recent years, important progress has been made in the lithologic trap exploration of the Lower Cretaceous Yageliemu Formation in the Yakela fault convex and its surrounding areas in the Tarim Basin, in order to clarify the sedimentary facies distribution law of the clastic rock reservoir of Yageliemu Formation in this area, and promote the efficient exploration and development of the clastic rock reservoir. Based on an integrated analysis of core samples, well logging data, and 3D seismic surveys, the sedimentary facies types and spatial distribution patterns of the Cretaceous Yageliemu Formation in the Yakela fault convex and its surrounding areas in the Tarim Basin were investigated. Furthermore, under the framework of source-to-sink system theory, the controlling effects of source area characteristics on fan delta development were systematically examined. The results show that fan delta group deposits are developed in the Yageliemu Formation in the Yakela fault convex and its surrounding areas. During the deposition period of the Yageliemu Formation, the ancient uplift in the Yakela fault convex area was obviously segmented, with a banded uplift in the NEE direction, with two bulges in the east and west, and a low terrain in the middle. Based on the analysis of source-sink system, it is clear that the sediment source of the fan delta group is from the weathered denudation area of the ancient uplift, and the multi-branch ancient gullies provide sediment transport channels for the fan delta Group. It can be seen that paleogeomorphology and gully development characteristics control the sour-sink system of Yageliemu formation in the Yakela fault convex and its surrounding areas, forming a sedimentary pattern with multiple sources supply. The western ancient uplift is mainly characterized by high uplift and large gully area,the eastern section is mainly characterized by low uplift and small gully area. The development scale of gullies in provenance area controlled the distribution scale of deltaic sediments around ancient uplift. The gentle slope fan delta sedimentary system developed in the south of the ancient uplift, and the steep slope and gentle slope fan delta sedimentary system developed in the north. The analysis of source-sink system reduces the uncertainty of sedimentary facies study of Yageliemu Formation and can provide more geological basis for oil and gas exploration.

In order to address a series of safety management issues involved in low-altitude economic development, the technical routes and principles of low-altitude economy as well as the operational experience of implementation plans are summarized, and four universal construction plans for low-altitude security and protection are analyzed, namely the radar and integrated perception technology fusion plan, the broadcast automatic dependent surveillance technology plan, the remote identification technology plan, and the multi-source fusion plan based on TDOA radio technology. On this basis, an evaluation index system for unmanned aerial vehicle detection technology was constructed, and a multi-attribute evaluation method based on DEMATEL and TOPSIS was established. The results show that the multi-source fusion plan based on time difference of arrival (TDOA) is an effective and universal solution for building a low-altitude security system in cities. It is concluded that the construction of a low-altitude security system is a systematic project, which requires the joint efforts of governments, enterprises and the whole society. Integration is needed at the levels of technology, data and operation to meet the future development needs of the low-altitude economy.

In order to do a good job of emergency management in the airport sector, strengthen the construction of the emergency response system, and improve the emergency response capability, a decision support method based on hybrid reasoning was proposed for the disposal of airport emergencies. Firstly, the ontology model of airport emergencies was constructed by abstracting the emergencies and the disposal process in the airport based on the actual scenarios and official documents. Secondly, the hybrid reasoning combining rule-based reasoning and case-based reasoning was introduced for case retrieval, and case representation was performed for the constructed ontology model to construct a case database. Lastly, the retrieval results are corrected using a feature weighting algorithm for attribute trade-offs, and the attribute parameters were adjusted using a neural network-based weight parameter optimization strategy. The advantages of the Bert+LSTM combination in this task scenario were verified by comparing it with commonly used deep learning models, and the final example proves that when an emergency occurs, the model can focus on the emergency itself, refer to historical cases and disposal standards, and obtain a structured data that comprehensively describes the information and disposal measures of the emergency, which provides support for emergency disposal decision making.